Historical Bias-corrected Surface Climate Dataset Based on Dynamically Downscaled CMIP6 Climate Projections for North America

This dataset provides an ensemble of dynamically downscaled and bias-corrected surface climate projections for North America, designed for hydrological modelling and climate change impact assessment. It was developed as part of the Canada1Water (C1W) initiative, which aims to support physically based modelling of climate change impacts on surface and groundwater systems across Canadian and transboundary basins. The dataset is based on three CMIP6 global climate projections (CESM2, MPI-ESM1.2-HR, MRI-ESM2.0) that were dynamically downscaled using the Weather Research and Forecasting (WRF) model with three different physics configurations, resulting in a nine-member ensemble. Here, only data for the historical period (1979–2015) are provided; data for a mid-century (2039–2060) and an end-century (2079–2100) time periods under the SSP3-7.0 greenhouse gas (GHG) concentration scenario are available here. The downscaled climate variables were bias-corrected using a gridded reference dataset derived from the Canadian Surface Reanalysis (CaSR) and AgERA5. Precipitation was corrected using the Local Intensity Scaling (LOCI) method, while other variables were corrected using two-moment linear scaling. The dataset has been generated at daily timesteps and 0.1° resolution across North America; however, due to data volume constraints, only monthly averages are available for download (daily data can be made available upon request). The dataset includes variables required for hydrological modelling, such as minimum and maximum air temperature, precipitation, humidity, wind speed, and downwelling shortwave and longwave radiation. In addition to the bias-corrected climate variables, the dataset includes derived hydrological forcing variables. Potential evapotranspiration (PET) was computed using the Penman–Monteith formulation, and snow water equivalent (SWE) and snowmelt were generated using a process-based snow model driven by the bias-corrected climate dataset. Bias-corrected soil temperature fields are also provided. The dataset supports a wide range of applications, including hydrological modelling, water resource assessment, snow and permafrost studies, and climate change impact analysis. A companion dataset with the raw WRF simulations output without bias correction is available here. The monthly normals of this dataset are also available through the Canada1Water Data Portal.